| *** jdg has quit IRC | 00:01 | |
| danwent | rohit404: hey | 00:16 |
|---|---|---|
| rohit404 | danwent: referring to the authN discussion, should we try to put in authentication support in Quantum manager then ? | 00:18 |
| danwent | rohit404: my thinking is that we should do as much as possible in the python-quantumclient lib, but yes, I suspect that nova will need to have some kind of flag to specify a keystone credential used to talk to nova. | 00:19 |
| danwent | sorry, nova -> quantum | 00:20 |
| danwent | btw, we'll be having a session on nova + quantum integration at the Folsom summit, but "QuantumManager" will likely go away in folsom, with nova-compute contacting Quantum directly | 00:22 |
| rohit404 | danwent: ah, i was trying to phrase my question to that only | 00:22 |
| *** heckj has quit IRC | 00:23 | |
| rohit404 | so like you mentioned, if nova-compute contacts quantum directly, we would still need some form of authentication | 00:24 |
| *** fattarsi has quit IRC | 00:24 | |
| *** roge has joined #openstack-dev | 00:32 | |
| rohit404 | danwent: i'll coordinate with Deepak and Maru and pitch in wherever I can for the python-quantumclient....i think the next logical step for keystone integration with Quantum would be AuthZ, right ? | 00:32 |
| *** roge has quit IRC | 00:40 | |
| *** mszilagyi has quit IRC | 00:40 | |
| rohit404 | danwent: btw, sorry for the delay in the quantum client checkin. had some issues with my env (fixed it now). have a good weekend. | 00:41 |
| *** maplebed has quit IRC | 00:49 | |
| *** shevek_ has joined #openstack-dev | 00:54 | |
| *** vincentricci has quit IRC | 00:59 | |
| *** jog0 has left #openstack-dev | 01:05 | |
| danwent | rohit404: yup, auth is needed regardless. Agree that Authz is main area, coordinating with Deepak would be great. Talk to you soon, have a good weekend | 01:05 |
| *** jdurgin has quit IRC | 01:06 | |
| *** johnpostlethwait has quit IRC | 01:06 | |
| *** kindaopsdevy has quit IRC | 01:06 | |
| *** littleidea has quit IRC | 01:19 | |
| *** misheska has quit IRC | 01:22 | |
| *** mnewby has quit IRC | 01:48 | |
| *** danwent has quit IRC | 01:57 | |
| *** devananda has quit IRC | 01:57 | |
| *** shevek_ has quit IRC | 01:59 | |
| *** spiffxp has quit IRC | 02:20 | |
| *** adjohn has quit IRC | 02:23 | |
| *** dolphm has joined #openstack-dev | 02:25 | |
| *** rods has quit IRC | 02:29 | |
| *** adjohn has joined #openstack-dev | 02:43 | |
| *** dolphm has quit IRC | 02:47 | |
| *** mdrnstm has quit IRC | 02:53 | |
| *** adjohn has quit IRC | 02:59 | |
| *** devcamcar has quit IRC | 03:48 | |
| *** zaitcev has quit IRC | 04:01 | |
| *** littleidea has joined #openstack-dev | 04:01 | |
| *** mnewby has joined #openstack-dev | 04:02 | |
| *** asalkeld has joined #openstack-dev | 04:05 | |
| *** asalkeld has left #openstack-dev | 04:05 | |
| *** vishy has joined #openstack-dev | 04:12 | |
| *** vishy has left #openstack-dev | 04:12 | |
| *** vishy has joined #openstack-dev | 04:12 | |
| *** mdrnstm has joined #openstack-dev | 04:17 | |
| *** lloydde has joined #openstack-dev | 04:28 | |
| *** adjohn has joined #openstack-dev | 04:56 | |
| *** adjohn has quit IRC | 04:58 | |
| *** hattwick has quit IRC | 05:02 | |
| *** RobertLaptop has joined #openstack-dev | 05:06 | |
| *** RobertLaptop has left #openstack-dev | 05:07 | |
| *** RobertLaptop_ has joined #openstack-dev | 05:15 | |
| *** RobertLaptop_ has left #openstack-dev | 05:15 | |
| *** RobertLaptop has joined #openstack-dev | 05:16 | |
| *** RobertLaptop has left #openstack-dev | 05:17 | |
| *** adjohn has joined #openstack-dev | 05:18 | |
| *** lloydde has quit IRC | 05:20 | |
| *** paulstallworth has quit IRC | 05:22 | |
| *** anotherjesse_zz is now known as anotherjesse | 05:29 | |
| *** danwent has joined #openstack-dev | 05:36 | |
| openstackgerrit | Verification of a change to openstack-dev/devstack failed: Allow skipping exercises. https://review.openstack.org/5428 | 05:45 |
| openstackgerrit | Verification of a change to openstack/python-keystoneclient failed: pep8 fixes for setup.py https://review.openstack.org/5639 | 05:47 |
| *** thingee has joined #openstack-dev | 05:59 | |
| *** dolphm has joined #openstack-dev | 06:04 | |
| *** dolphm has quit IRC | 06:16 | |
| *** bepernoot has joined #openstack-dev | 06:19 | |
| *** anotherjesse is now known as anotherjesse_zz | 06:26 | |
| *** danwent has quit IRC | 06:28 | |
| *** sandywalsh has quit IRC | 06:54 | |
| *** reed has quit IRC | 06:58 | |
| *** thingee has quit IRC | 07:03 | |
| *** bepernoot has quit IRC | 07:15 | |
| *** rohit404 has quit IRC | 07:27 | |
| *** sandywalsh has joined #openstack-dev | 07:36 | |
| *** adjohn has quit IRC | 07:42 | |
| *** relateable has joined #openstack-dev | 07:57 | |
| *** hattwick has joined #openstack-dev | 08:00 | |
| *** mnewby has quit IRC | 08:20 | |
| *** mnewby has joined #openstack-dev | 08:22 | |
| *** mnewby has quit IRC | 08:23 | |
| *** littleidea has quit IRC | 08:25 | |
| *** bepernoot has joined #openstack-dev | 08:38 | |
| *** bepernoot has quit IRC | 08:42 | |
| *** bepernoot has joined #openstack-dev | 08:47 | |
| *** shevek_ has joined #openstack-dev | 08:54 | |
| *** dachary has joined #openstack-dev | 09:16 | |
| *** BruceZu-CNBJ has joined #openstack-dev | 09:18 | |
| *** bepernoot has quit IRC | 09:19 | |
| *** bepernoot has joined #openstack-dev | 09:20 | |
| *** bepernoot has quit IRC | 09:28 | |
| *** tserong has quit IRC | 09:29 | |
| *** tserong has joined #openstack-dev | 09:30 | |
| *** tserong has joined #openstack-dev | 09:30 | |
| *** zigo has joined #openstack-dev | 09:56 | |
| *** relateable has quit IRC | 09:56 | |
| *** deepakgarg has joined #openstack-dev | 12:10 | |
| *** deepakgarg has quit IRC | 12:22 | |
| *** dachary has quit IRC | 13:47 | |
| *** littleidea has joined #openstack-dev | 13:55 | |
| *** ayoung has quit IRC | 14:02 | |
| *** flaviamissi has joined #openstack-dev | 14:04 | |
| *** dachary has joined #openstack-dev | 14:14 | |
| *** paulstallworth has joined #openstack-dev | 14:22 | |
| *** dolphm has joined #openstack-dev | 14:22 | |
| *** dolphm has quit IRC | 14:39 | |
| *** dtroyer_zzz is now known as dtroyer | 14:45 | |
| *** zigo has quit IRC | 15:04 | |
| *** dolphm has joined #openstack-dev | 15:14 | |
| *** dolphm has quit IRC | 15:21 | |
| *** clayg is now known as 65MAADRBZ | 15:43 | |
| *** zigo has joined #openstack-dev | 15:43 | |
| *** glenc has joined #openstack-dev | 15:43 | |
| *** mtaylor has joined #openstack-dev | 15:43 | |
| *** clayg has joined #openstack-dev | 15:43 | |
| *** troytoman-away has joined #openstack-dev | 15:43 | |
| *** hubbard.freenode.net sets mode: +ov mtaylor mtaylor | 15:43 | |
| *** sdake has quit IRC | 15:43 | |
| *** littleidea has quit IRC | 15:48 | |
| *** dtroyer is now known as dtroyer_zzz | 15:58 | |
| *** ywu has joined #openstack-dev | 16:01 | |
| *** bepernoot has joined #openstack-dev | 16:02 | |
| *** bepernoot has quit IRC | 16:05 | |
| *** zigo has quit IRC | 16:06 | |
| *** zigo has joined #openstack-dev | 16:06 | |
| *** novas0x2a|laptop has joined #openstack-dev | 16:27 | |
| *** roge has joined #openstack-dev | 16:31 | |
| *** dolphm has joined #openstack-dev | 16:35 | |
| *** maplebed has joined #openstack-dev | 16:37 | |
| *** dolphm has quit IRC | 16:39 | |
| *** roge has quit IRC | 16:42 | |
| *** dtroyer_zzz is now known as dtroyer | 16:51 | |
| *** danwent has joined #openstack-dev | 16:51 | |
| *** dolphm has joined #openstack-dev | 16:56 | |
| *** devananda has joined #openstack-dev | 17:34 | |
| *** novas0x2a|laptop has quit IRC | 17:41 | |
| *** dachary has quit IRC | 17:44 | |
| *** maplebed_ has joined #openstack-dev | 17:48 | |
| *** maplebed has quit IRC | 17:48 | |
| *** maplebed_ is now known as Guest22135 | 17:48 | |
| *** Guest22135 has quit IRC | 17:51 | |
| *** Guest22135 has joined #openstack-dev | 17:51 | |
| *** Guest22135 is now known as maplebed | 17:51 | |
| *** dtroyer is now known as dtroyer_zzz | 17:51 | |
| *** roge has joined #openstack-dev | 17:58 | |
| *** roge has quit IRC | 18:01 | |
| *** BruceZu-CNBJ has quit IRC | 18:06 | |
| *** dachary has joined #openstack-dev | 18:07 | |
| *** BruceZu-home has joined #openstack-dev | 18:07 | |
| *** danwent has quit IRC | 18:07 | |
| *** BruceZu-home has quit IRC | 18:08 | |
| *** zigo-_- has joined #openstack-dev | 18:08 | |
| *** zigo has quit IRC | 18:09 | |
| *** winston-d has joined #openstack-dev | 18:11 | |
| *** relateable has joined #openstack-dev | 18:13 | |
| winston-d | hi guys, can someone help and review my implementation of multiprocess Nova API services? https://review.openstack.org/#change,5762 | 18:13 |
| *** devananda has quit IRC | 18:18 | |
| *** dolphm has quit IRC | 18:24 | |
| *** winston-d has quit IRC | 18:25 | |
| *** littleidea has joined #openstack-dev | 18:25 | |
| *** bepernoot has joined #openstack-dev | 18:27 | |
| *** mszilagyi has joined #openstack-dev | 18:34 | |
| *** eglynn has joined #openstack-dev | 18:34 | |
| *** relateable has quit IRC | 18:35 | |
| *** mszilagyi has quit IRC | 18:36 | |
| *** sdake has joined #openstack-dev | 18:40 | |
| *** ncode has joined #openstack-dev | 18:41 | |
| *** ncode has joined #openstack-dev | 18:41 | |
| *** maploin has joined #openstack-dev | 18:43 | |
| *** maploin has quit IRC | 18:43 | |
| *** maploin has joined #openstack-dev | 18:43 | |
| *** ncode has quit IRC | 18:44 | |
| *** sdake has quit IRC | 18:51 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:52 | |
| *** bepernoot has joined #openstack-dev | 19:10 | |
| *** thingee has joined #openstack-dev | 19:11 | |
| *** bepernoot has quit IRC | 19:22 | |
| *** flaviamissi has quit IRC | 19:30 | |
| *** flaviamissi has joined #openstack-dev | 19:31 | |
| *** littleidea has quit IRC | 19:31 | |
| *** littleidea has joined #openstack-dev | 19:33 | |
| *** littleidea has quit IRC | 19:42 | |
| *** flaviamissi has quit IRC | 19:47 | |
| *** flaviamissi has joined #openstack-dev | 19:48 | |
| *** zigo-_- has quit IRC | 19:56 | |
| *** reed has joined #openstack-dev | 19:58 | |
| *** dolphm has joined #openstack-dev | 19:58 | |
| *** maploin has quit IRC | 20:17 | |
| *** roge has joined #openstack-dev | 20:21 | |
| *** flaviamissi_ has joined #openstack-dev | 20:25 | |
| *** flaviamissi has quit IRC | 20:28 | |
| *** flaviamissi_ has quit IRC | 20:29 | |
| *** flaviamissi has joined #openstack-dev | 20:32 | |
| *** flaviamissi has quit IRC | 20:37 | |
| *** flaviamissi has joined #openstack-dev | 20:38 | |
| *** johnpostlethwait has joined #openstack-dev | 20:43 | |
| *** flaviamissi has quit IRC | 20:43 | |
| *** flaviamissi has joined #openstack-dev | 20:43 | |
| *** novas0x2a|laptop has quit IRC | 20:45 | |
| *** roge has quit IRC | 20:46 | |
| *** flaviamissi has quit IRC | 20:47 | |
| *** flaviamissi has joined #openstack-dev | 20:48 | |
| *** flaviamissi has quit IRC | 20:50 | |
| *** dtroyer_zzz is now known as dtroyer | 20:52 | |
| *** maplebed_ has joined #openstack-dev | 20:54 | |
| *** maplebed_ is now known as Guest42149 | 20:55 | |
| *** dolphm has quit IRC | 20:55 | |
| *** maplebed has quit IRC | 20:57 | |
| *** dtroyer is now known as dtroyer_zzz | 21:00 | |
| *** Guest42149 has quit IRC | 21:19 | |
| *** dtroyer_zzz is now known as dtroyer | 21:20 | |
| *** rods has joined #openstack-dev | 21:30 | |
| *** roge has joined #openstack-dev | 21:31 | |
| *** anotherjesse_zz is now known as anotherjesse | 21:47 | |
| *** novas0x2a|laptop has joined #openstack-dev | 21:52 | |
| *** roge has quit IRC | 22:08 | |
| *** Shrews has quit IRC | 22:08 | |
| *** Ahmad_HP has joined #openstack-dev | 22:10 | |
| Ahmad_HP | Hi Vish | 22:10 |
| Ahmad_HP | there? | 22:11 |
| Ahmad_HP | @vishy | 22:14 |
| *** anotherjesse is now known as anotherjesse_zz | 22:21 | |
| vishy | y | 22:27 |
| *** sleepsonthefloor is now known as sleepsonzzz | 22:29 | |
| *** lloydde has joined #openstack-dev | 22:29 | |
| *** danwent has joined #openstack-dev | 22:38 | |
| Ahmad_HP | just have a quick query | 22:44 |
| Ahmad_HP | was this ever implemented: https://blueprints.launchpad.net/nova/+spec/nova-metadata-nodb | 22:44 |
| Ahmad_HP | @vishy | 22:48 |
| vishy | no | 22:48 |
| vishy | although it caches the request fo 15 seconds | 22:48 |
| vishy | * for | 22:49 |
| vishy | so it is much faster now | 22:49 |
| Ahmad_HP | hmmmm... | 22:49 |
| Ahmad_HP | I am checking for the possiblity if we even serve the first request through cache | 22:49 |
| Ahmad_HP | which means, generating cache on instance creation request time | 22:50 |
| Ahmad_HP | @Vishy, Thanks by the way | 22:55 |
| vishy | Ahmad_HP: that would work as long as you are expecting the data to never change | 22:56 |
| vishy | Ahmad_HP: if you know it won't change, you could up the 15 second cache to forever | 22:57 |
| vishy | Ahmad_HP: and send a request at boot time to generate it | 22:57 |
| vishy | although I don't know what advantage that gives you performance wise | 22:58 |
| vishy | save a few hundred ms on the first request? | 22:58 |
| Ahmad_HP | But the problem of hitting the first request to db is that, if db request takes a bit longer then cloud init times out and send request again | 22:58 |
| Ahmad_HP | we are seeing scenarios where metadata requests are kind of overloading the whole system. | 22:59 |
| Ahmad_HP | because metadata retrival is not trivial | 23:00 |
| Ahmad_HP | retrieval* | 23:00 |
| vishy | Ahmad_HP: how long is the first request taking? | 23:01 |
| Ahmad_HP | even with cache........... the longer time in first metadata requests causes cloud iniit to time out and eventually making requests multiple times | 23:01 |
| vishy | well you could have a simple call in compute to prepopulate the cache | 23:01 |
| vishy | if you would like | 23:01 |
| vishy | it could even import the metadata code and just run it | 23:02 |
| vishy | if you want to get fancy, you could do it with a periodic task | 23:02 |
| vishy | and make the memcached timeout longer than the timeout | 23:02 |
| vishy | that way you could handle it changing as well | 23:03 |
| Ahmad_HP | It depends on the system load........ somes cases are even taking 70 to 200 seconds....these are worst ones | 23:03 |
| vishy | wow | 23:03 |
| vishy | sounds like you need to optimize the db! | 23:04 |
| *** ywu_ has joined #openstack-dev | 23:04 | |
| Ahmad_HP | true | 23:04 |
| vishy | Ahmad_HP: why not have a read slave just for metadata | 23:04 |
| * dachary reading http://nova.openstack.org/devref/zone.html | 23:04 | |
| vishy | it never has to write | 23:04 |
| vishy | you could have the metadata server always hit the readslave | 23:04 |
| Ahmad_HP | interesting idea | 23:04 |
| vishy | seems like the easiest way to deal with the performance issue | 23:05 |
| *** ywu has quit IRC | 23:06 | |
| Ahmad_HP | don't know how much performance will be affected by just synching the read slave with master | 23:06 |
| Ahmad_HP | I hate seeing even a single time out in cloud init metadata request............I want to address this | 23:08 |
| Ahmad_HP | 1) Either trigger preloading metadata in compute node | 23:16 |
| Ahmad_HP | 2) If main API servers already have this information during insteance creation time, then why not use this to populate cache | 23:17 |
| *** roge has joined #openstack-dev | 23:18 | |
| *** novas0x2a|lapto1 has joined #openstack-dev | 23:26 | |
| *** novas0x2a|laptop has quit IRC | 23:27 | |
| *** roge has quit IRC | 23:29 | |
| *** dolphm has joined #openstack-dev | 23:31 | |
| *** kindaopsdevy has joined #openstack-dev | 23:34 | |
| *** kindaopsdevy has quit IRC | 23:42 | |
| *** novas0x2a|lapto1 has quit IRC | 23:44 | |
| *** maplebed has joined #openstack-dev | 23:46 | |
| *** kindaopsdevy has joined #openstack-dev | 23:47 | |
| *** dolphm has quit IRC | 23:48 | |
| *** dolphm has joined #openstack-dev | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!